module

ML::LLM

Class methods

cleanup

Cleanup llama backend (call at program end)

gpu_available?

Check if GPU offload is supported

init

Initialize llama backend (call once at program start)

register_context(context : Context) : Nil

register_model(model : Model) : Nil

system_info

Get system info string

unregister_context(context : Context) : Nil

unregister_model(model : Model) : Nil